The official Node SDK now accepts grok_imagine_1_5_lite for text-to-video and image-to-video. Upgrade the package and swap the model.
Version 4.21.0 of Runway's official Node SDK, released October 2, 2026, adds a new video model, Grok Imagine Video 1.5 Lite (grok_imagine_1_5_lite), to both text-to-video and image-to-video. The same release adds the Eleven v4 text-to-speech model.
Wirestack
RUNWAYML_API_SECRETnpm install @runwayml/sdk@4.21.0import RunwayML from '@runwayml/sdk';
const client = new RunwayML({
apiKey: process.env['RUNWAYML_API_SECRET'],
});
const imageToVideo = await client.imageToVideo.create({
model: 'gen4_turbo',
promptImage: 'https://example.com/assets/bunny.jpg',
ratio: '1280:720',
promptText: 'The bunny is eating a carrot',
});
console.log(imageToVideo.id);model to 'grok_imagine_1_5_lite'. The SDK changelog says it is accepted for text-to-video as well, so you can start from a prompt with no image.Run the script. It prints a task ID; the finished clip is fetched with that ID, as in the Wirestack recipe Generate a 5-second AI video clip from a prompt with the Runway API.
ratio and duration values the new model accepts. The example's 1280:720 is the value Runway shows for gen4_turbo; check the API reference before relying on it.